Certified Data Removal from Machine Learning Models

By Chuan Guo, Tom Goldstein, Awni Y. Hannun, L. V. D. van der Maaten. 2019

687 citations recorded by the source corpus.

Good data stewardship requires removal of data at the request of the data's owner. This raises the question if and how a trained machine-learning model, which implicitly stores information about its training data, should be affected by such a removal request. Is it possible to "remove" data from a machine-learning model? We study this problem by defining certified removal: a very strong theoretical guarantee that a model from which data is removed cannot be distinguished from a model that never observed the data to begin with. We develop a certified-removal mechanism for linear classifiers and empirically study learning settings in which this mechanism is practical.

Subfields: Safety, Robustness & Interpretability, Privacy & Security

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Safety, Robustness & Interpretability

In Privacy & Security

Citations in the corpus

Mixed-Privacy Forgetting in Deep Networks (2020, methodology): “Can we forget the cohort of interest without interfering with information about the other data and preserving, to the extent possible, the accuracy of the trained model? Recently, the problem of forgetting has received considerable attention [15, 16, 13, 19, 5, 24, 35, 41, 6, 39, 12, 7, 36], but sol”
DeltaGrad: Rapid retraining of machine learning models (2020, background): “Such ideas have been used recently e.g., Ginart et al. (2019); Guo et al. (2019); Wu et al. (2020).”
Machine Unlearning (2019, background): “This includes two concurrent proposals [34], [35] The mechanism by Guo et al.”
Measuring Forgetting of Memorized Training Examples (2022, neutral): “Algorithms for machine unlearning have been proposed for k-means [24], empirical risk minimization [29, 35, 59, 76, 67], and deep learning [18, 26, 25, 60, 5].”
LQF: Linear Quadratic Fine-Tuning (2020, methodology): “We leave various applications of LQF to future work, including its use in (i) uncertainty quantification, (ii) backward compatibility [44], (iii) leave-one-out cross validation without a validation set [9], (iv) active/online learning [45], (v) counterfactual analysis in the first order, (vi) contin”
Approximate Data Deletion from Machine Learning Models: Algorithms and Evaluations (2020, methodology): “Newton’s method Recent work (Guo et al., 2019) has attempted approximate retraining by taking a single step of Newton’s method.”
DART: Data Addition and Removal Trees (2020, neutral): “2019), SVMs (Cauwenberghs and Poggio 2001), logistic regression (Guo et al. 2020; Schelter 2020), and deep neural networks (Baumhauer, Schöttle, and Zeppelzauer 2020; Golatkar, Achille, and Soatto 2020b; Wu, Dobriban, and Davidson 2020) (for a comprehensive list of related work, see Section 6); howe”
When Machine Unlearning Jeopardizes Privacy (2020, background): “This process, also called machine unlearning, has gained momentum both in academia and industry [7, 9, 10, 17, 18, 20, 27, 35, 50, 55, 62].”
Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning (2024, methodology): “Since its proposal by Cao & Yang (2015), machine unlearning has been extensively studied in the classification literature (Bourtoule et al., 2021; Golatkar et al., 2020; Ginart et al., 2019; Thudi et al., 2022; Izzo et al., 2021; Koh & Liang, 2017; Guo et al., 2019; Sekhari et al., 2021).”
Machine Unlearning for Random Forests (2020, methodology): “, 2019), SVMs (Cauwenberghs & Poggio, 2001), logistic regression (Guo et al., 2020; Schelter, 2020), and deep neural networks (Baumhauer et al.”
Variational Bayesian Unlearning (2020, background): “[14] C.”
Fast Yet Effective Machine Unlearning (2021, methodology): “[23] give a certified information removal framework based on Newton’s update removal mechanism for convex learning problems.”
Unrolling SGD: Understanding Factors Influencing Machine Unlearning (2021, methodology): “Instead of retraining, these methods execute computationally less expensive operations on the final weights [11, 8, 10], apply some architectural change [12], or filter outputs [12].”
Amnesiac Machine Learning (2020, methodology): “The work presented in Ginart et al. (Gi-nart et al. 2019) and Guo et al. (Guo et al. 2019) are both focused on such different machine learning methods that it is impossible to make a direct comparison with our work.”
Large Language Model Unlearning (2023, background): “…for classification models, including data-reversed training [Tarun et al., 2023, Liu et al., 2022, Chundawat et al., 2023], optimization-based unlearning [Guo et al., 2019, Neel et al., 2021] and influence function based approaches [Jia et al., 2023, Warnecke et al., 2021, Izzo et al., 2021].”
Towards Probabilistic Verification of Machine Unlearning (2020, methodology): “Second, the server can use our backdoor-based mechanism to quantitatively measure the effectiveness of recently proposed deletion approaches without strict deletion guarantees, such as [10, 11].”